From fd8309493307c92d0ba45e139157aabe81859c31 Mon Sep 17 00:00:00 2001 From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com> Date: Fri, 14 Aug 2026 12:10:37 +0000 Subject: [PATCH 1/2] docs: clarify API billing vs consumer plans and reasoning model discovery --- guides/features/reasoning-models.mdx | 2 ++ overview/pricing.mdx | 12 ++++++++++++ 2 files changed, 14 insertions(+) diff --git a/guides/features/reasoning-models.mdx b/guides/features/reasoning-models.mdx index b42c3885..a58736c6 100644 --- a/guides/features/reasoning-models.mdx +++ b/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ When in doubt, use `low`, `medium`, or `high`. These are the most widely support ### Model support +The tables below list models with known `reasoning_effort` support. Other models may still generate reasoning tokens (visible in `usage.completion_tokens_details.reasoning_tokens`) without exposing an effort control. To check whether a specific model supports reasoning or the `reasoning_effort` parameter, read the `supportsReasoning` and `supportsReasoningEffort` fields on the [`/v1/models`](/api-reference/endpoint/models/list) endpoint. + #### OpenAI | Model | Supported values | diff --git a/overview/pricing.mdx b/overview/pricing.mdx index 8bfee953..2df2ea01 100644 --- a/overview/pricing.mdx +++ b/overview/pricing.mdx @@ -6,6 +6,16 @@ title: API Pricing Prices per 1M tokens unless noted. All prices in USD. 1 Diem = $1/day of compute. +## How the API bills + +The Venice API is **pay-as-you-go**: every request is metered and debited from your account's Venice credit balance (funded in USD or via staked DIEM). There is no free tier, no monthly API allowance, and no login-based subscription plan for the API — access is authenticated by API key and billed per request. + + +The Free, Pro, and Max plans on the [venice.ai](https://venice.ai) consumer apps (web, desktop, and mobile chat) are separate from the API. Consumer subscription allowances — image counts, character quotas, unlimited chat, and so on — apply only to those apps and do **not** apply to API usage. API calls are always metered from your Venice credit balance regardless of your consumer plan. See [Pro Users](#pro-users) below for the one-time credit Pro subscribers receive. + + +To use the API, you need an [API key](/guides/getting-started/generating-api-key) and available balance, funded through one of the [Payment Options](#payment-options) below. Every model listed on this page is available to any API key with sufficient balance — there is no separate "base" versus "Pro" model split on the API. + ## Text Models ### Chat Completions @@ -478,3 +488,5 @@ These charges apply in addition to standard model token pricing. ### Pro Users Pro subscribers receive a one-time $10 Venice credit when upgrading to Pro. Use it to test and build small apps. + +Beyond that credit, API usage continues to be billed at the same pay-as-you-go rates listed above — a Pro subscription does not include additional API allowances such as bundled images, characters, or tokens. Some models are restricted to Pro accounts (`PRO_ONLY_MODEL`); calling them from an account without an active Pro subscription returns `AUTHENTICATION_FAILED_INACTIVE_KEY`. See [Error Codes](/api-reference/error-codes). From c3167b79fc67c0a3e3003f815cf164408232409c Mon Sep 17 00:00:00 2001 From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com> Date: Fri, 14 Aug 2026 12:14:57 +0000 Subject: [PATCH 2/2] docs: translate reasoning-models note, improve pricing SEO, split em-dashes --- ar/guides/features/reasoning-models.mdx | 2 ++ de/guides/features/reasoning-models.mdx | 2 ++ es/guides/features/reasoning-models.mdx | 2 ++ fr/guides/features/reasoning-models.mdx | 2 ++ it/guides/features/reasoning-models.mdx | 2 ++ ko/guides/features/reasoning-models.mdx | 2 ++ overview/pricing.mdx | 13 +++++++------ pt-BR/guides/features/reasoning-models.mdx | 2 ++ zh/guides/features/reasoning-models.mdx | 2 ++ 9 files changed, 23 insertions(+), 6 deletions(-) diff --git a/ar/guides/features/reasoning-models.mdx b/ar/guides/features/reasoning-models.mdx index bf2ea47f..6a7d7ccd 100644 --- a/ar/guides/features/reasoning-models.mdx +++ b/ar/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ curl https://api.venice.ai/api/v1/chat/completions \ ### دعم النموذج +تسرد الجداول أدناه النماذج التي يُعرف دعمها لـ `reasoning_effort`. قد تظل نماذج أخرى تولّد توكنات استدلال (تظهر في `usage.completion_tokens_details.reasoning_tokens`) دون أن توفر تحكمًا في مستوى الجهد. للتحقق مما إذا كان نموذج معيّن يدعم الاستدلال أو معلمة `reasoning_effort`، اقرأ حقلَي `supportsReasoning` و`supportsReasoningEffort` في نقطة النهاية [`/v1/models`](/api-reference/endpoint/models/list). + #### OpenAI | النموذج | القيم المدعومة | diff --git a/de/guides/features/reasoning-models.mdx b/de/guides/features/reasoning-models.mdx index c8dad729..a03e298b 100644 --- a/de/guides/features/reasoning-models.mdx +++ b/de/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ Verwenden Sie im Zweifel `low`, `medium` oder `high`. Dies sind die am breiteste ### Modellunterstützung +Die folgenden Tabellen listen Modelle mit bekannter Unterstützung für `reasoning_effort` auf. Andere Modelle können weiterhin Reasoning-Tokens erzeugen (sichtbar in `usage.completion_tokens_details.reasoning_tokens`), ohne eine Effort-Steuerung anzubieten. Um zu prüfen, ob ein bestimmtes Modell Reasoning oder den Parameter `reasoning_effort` unterstützt, lesen Sie die Felder `supportsReasoning` und `supportsReasoningEffort` am Endpoint [`/v1/models`](/api-reference/endpoint/models/list). + #### OpenAI | Modell | Unterstützte Werte | diff --git a/es/guides/features/reasoning-models.mdx b/es/guides/features/reasoning-models.mdx index a558681c..ed776374 100644 --- a/es/guides/features/reasoning-models.mdx +++ b/es/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ En caso de duda, usa `low`, `medium` o `high`. Son los valores más ampliamente ### Soporte por modelo +Las tablas siguientes enumeran los modelos con soporte conocido para `reasoning_effort`. Otros modelos pueden seguir generando tokens de razonamiento (visibles en `usage.completion_tokens_details.reasoning_tokens`) sin exponer un control de esfuerzo. Para comprobar si un modelo concreto admite razonamiento o el parámetro `reasoning_effort`, lee los campos `supportsReasoning` y `supportsReasoningEffort` en el endpoint [`/v1/models`](/api-reference/endpoint/models/list). + #### OpenAI | Modelo | Valores admitidos | diff --git a/fr/guides/features/reasoning-models.mdx b/fr/guides/features/reasoning-models.mdx index e7179cf5..c69360e2 100644 --- a/fr/guides/features/reasoning-models.mdx +++ b/fr/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ En cas de doute, utilisez `low`, `medium` ou `high`. Ce sont les valeurs les plu ### Prise en charge des modèles +Les tableaux ci-dessous répertorient les modèles dont la prise en charge de `reasoning_effort` est connue. D'autres modèles peuvent tout de même générer des tokens de raisonnement (visibles dans `usage.completion_tokens_details.reasoning_tokens`) sans exposer de contrôle d'effort. Pour vérifier si un modèle donné prend en charge le raisonnement ou le paramètre `reasoning_effort`, lisez les champs `supportsReasoning` et `supportsReasoningEffort` sur l'endpoint [`/v1/models`](/api-reference/endpoint/models/list). + #### OpenAI | Modèle | Valeurs prises en charge | diff --git a/it/guides/features/reasoning-models.mdx b/it/guides/features/reasoning-models.mdx index ac05d5f7..5845e5d5 100644 --- a/it/guides/features/reasoning-models.mdx +++ b/it/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ In caso di dubbi, usa `low`, `medium` o `high`. Sono i valori più ampiamente su ### Supporto dei modelli +Le tabelle seguenti elencano i modelli con supporto noto a `reasoning_effort`. Altri modelli possono comunque generare token di ragionamento (visibili in `usage.completion_tokens_details.reasoning_tokens`) senza esporre un controllo dell'effort. Per verificare se un modello specifico supporta il ragionamento o il parametro `reasoning_effort`, leggi i campi `supportsReasoning` e `supportsReasoningEffort` sull'endpoint [`/v1/models`](/api-reference/endpoint/models/list). + #### OpenAI | Modello | Valori supportati | diff --git a/ko/guides/features/reasoning-models.mdx b/ko/guides/features/reasoning-models.mdx index 71fe4a55..e3c80315 100644 --- a/ko/guides/features/reasoning-models.mdx +++ b/ko/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ curl https://api.venice.ai/api/v1/chat/completions \ ### 모델 지원 +아래 표는 `reasoning_effort`를 지원하는 것으로 알려진 모델을 나열합니다. 다른 모델도 effort 제어를 노출하지 않은 채로 여전히 추론 토큰을 생성할 수 있으며(`usage.completion_tokens_details.reasoning_tokens`에서 확인 가능), 특정 모델이 추론이나 `reasoning_effort` 파라미터를 지원하는지 확인하려면 [`/v1/models`](/api-reference/endpoint/models/list) 엔드포인트의 `supportsReasoning` 및 `supportsReasoningEffort` 필드를 읽으세요. + #### OpenAI | Model | Supported values | diff --git a/overview/pricing.mdx b/overview/pricing.mdx index 2df2ea01..7924f7cd 100644 --- a/overview/pricing.mdx +++ b/overview/pricing.mdx @@ -1,20 +1,21 @@ --- -title: API Pricing +title: Venice API Pricing +description: "Venice API pricing per 1M tokens for text, image, video, audio, and embedding models, plus credit costs for web search, RPC calls, and other services." "og:title": "Pricing | Venice API Docs" -"og:description": "Learn about pricing for Venice's API." +"og:description": "Learn about Venice API pricing per 1M tokens across text, image, video, audio, and embedding models." --- Prices per 1M tokens unless noted. All prices in USD. 1 Diem = $1/day of compute. ## How the API bills -The Venice API is **pay-as-you-go**: every request is metered and debited from your account's Venice credit balance (funded in USD or via staked DIEM). There is no free tier, no monthly API allowance, and no login-based subscription plan for the API — access is authenticated by API key and billed per request. +The Venice API is **pay-as-you-go**: every request is metered and debited from your account's Venice credit balance (funded in USD or via staked DIEM). There is no free tier, no monthly API allowance, and no login-based subscription plan for the API. Access is authenticated by API key and billed per request. -The Free, Pro, and Max plans on the [venice.ai](https://venice.ai) consumer apps (web, desktop, and mobile chat) are separate from the API. Consumer subscription allowances — image counts, character quotas, unlimited chat, and so on — apply only to those apps and do **not** apply to API usage. API calls are always metered from your Venice credit balance regardless of your consumer plan. See [Pro Users](#pro-users) below for the one-time credit Pro subscribers receive. +The Free, Pro, and Max plans on the [venice.ai](https://venice.ai) consumer apps (web, desktop, and mobile chat) are separate from the API. Consumer subscription allowances such as image counts, character quotas, and unlimited chat apply only to those apps and do **not** apply to API usage. API calls are always metered from your Venice credit balance regardless of your consumer plan. See [Pro Users](#pro-users) below for the one-time credit Pro subscribers receive. -To use the API, you need an [API key](/guides/getting-started/generating-api-key) and available balance, funded through one of the [Payment Options](#payment-options) below. Every model listed on this page is available to any API key with sufficient balance — there is no separate "base" versus "Pro" model split on the API. +To use the API, you need an [API key](/guides/getting-started/generating-api-key) and available balance, funded through one of the [Payment Options](#payment-options) below. Every model listed on this page is available to any API key with sufficient balance. There is no separate "base" versus "Pro" model split on the API. ## Text Models @@ -489,4 +490,4 @@ These charges apply in addition to standard model token pricing. Pro subscribers receive a one-time $10 Venice credit when upgrading to Pro. Use it to test and build small apps. -Beyond that credit, API usage continues to be billed at the same pay-as-you-go rates listed above — a Pro subscription does not include additional API allowances such as bundled images, characters, or tokens. Some models are restricted to Pro accounts (`PRO_ONLY_MODEL`); calling them from an account without an active Pro subscription returns `AUTHENTICATION_FAILED_INACTIVE_KEY`. See [Error Codes](/api-reference/error-codes). +Beyond that credit, API usage continues to be billed at the same pay-as-you-go rates listed above. A Pro subscription does not include additional API allowances such as bundled images, characters, or tokens. Some models are restricted to Pro accounts (`PRO_ONLY_MODEL`); calling them from an account without an active Pro subscription returns `AUTHENTICATION_FAILED_INACTIVE_KEY`. See [Error Codes](/api-reference/error-codes). diff --git a/pt-BR/guides/features/reasoning-models.mdx b/pt-BR/guides/features/reasoning-models.mdx index b23e99cf..b2220fb6 100644 --- a/pt-BR/guides/features/reasoning-models.mdx +++ b/pt-BR/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ Em caso de dúvida, use `low`, `medium` ou `high`. Esses são os valores mais am ### Suporte por modelo +As tabelas abaixo listam os modelos com suporte conhecido a `reasoning_effort`. Outros modelos ainda podem gerar tokens de raciocínio (visíveis em `usage.completion_tokens_details.reasoning_tokens`) sem expor um controle de esforço. Para verificar se um modelo específico suporta raciocínio ou o parâmetro `reasoning_effort`, leia os campos `supportsReasoning` e `supportsReasoningEffort` no endpoint [`/v1/models`](/api-reference/endpoint/models/list). + #### OpenAI | Modelo | Valores suportados | diff --git a/zh/guides/features/reasoning-models.mdx b/zh/guides/features/reasoning-models.mdx index b71728e5..9050dcbb 100644 --- a/zh/guides/features/reasoning-models.mdx +++ b/zh/guides/features/reasoning-models.mdx @@ -124,6 +124,8 @@ curl https://api.venice.ai/api/v1/chat/completions \ ### 模型支持 +下表列出了已知支持 `reasoning_effort` 的模型。其他模型仍可能生成推理 token（可在 `usage.completion_tokens_details.reasoning_tokens` 中查看），但不提供 effort 控制。要查看特定模型是否支持推理或 `reasoning_effort` 参数，请读取 [`/v1/models`](/api-reference/endpoint/models/list) 端点上的 `supportsReasoning` 和 `supportsReasoningEffort` 字段。 + #### OpenAI | 模型 | 支持的值 |